There is a good chance your Shopify store is already selling inside ChatGPT right now, and you have no idea it is happening.
On March 24, 2026, Shopify switched on Agentic Storefronts for every eligible store. Overnight, products from millions of stores became discoverable, and in some cases buyable, inside ChatGPT, Microsoft Copilot, Google AI Mode, and the Gemini app. You did not install an app. You did not write a line of code. If you are on Shopify, you were opted in by default.
Here is why that matters. AI-driven traffic to Shopify stores is up about 7x since January 2025, and AI-attributed orders are up about 11x over the same window. ChatGPT alone has a reported 800 million weekly users. A real and growing slice of product discovery has moved out of Google and into AI conversations, and the stores that show up in those conversations with clean, structured product data are quietly capturing sales that would never have existed otherwise.
This is a long one, on purpose. I wrote it as a do-it-yourself guide so you can set this up and optimize it on your own, step by step. I would rather you actually understand how this works than hand you a vague summary. Let’s get into it.
What Are Shopify Agentic Storefronts?
Shopify Agentic Storefronts is a native feature that makes your products discoverable and purchasable across AI platforms, automatically, with no setup required. Your catalog gets pulled into AI shopping conversations so that when someone asks an AI assistant for a product, yours can show up as a recommendation and, on most platforms, be bought on the spot.
The engine behind it is the Shopify Catalog, a structured product database that uses specialized AI models to read, enrich, and standardize product data from millions of merchants. When Shopify announced this in the Winter 2026 Edition and then flipped it live for everyone on March 24, your products were syndicated into that catalog by default, unless you opted out.
The mental shift to make here is this: the AI assistant is the new front door, and your storefront becomes the checkout and fulfillment engine behind it. That does not make your store less important. As you will see at the end of this guide, it makes it more important.
How Selling Works on Each AI Platform
Not every AI platform sells the same way. Some complete the purchase right inside the chat. Others send the buyer to your own checkout. The protocol, the fees, and how much of the customer relationship you keep all change from platform to platform. Here is the honest breakdown of each one.
ChatGPT (the highest-reach channel)
- Protocol: the Agentic Commerce Protocol (ACP), built by OpenAI and Stripe.
- Where checkout happens: on your store. On mobile it opens in an in-app browser; on desktop it opens a new tab. The buyer lands on your real Shopify checkout.
- What you keep: your full checkout, upsells, payment methods, and customer data. ChatGPT is the discovery layer, and the sale closes in your house.
- Fees: standard Shopify processing rates, with no extra transaction fee in the live implementation. Earlier reports of a 4 percent OpenAI fee do not match what is live today, and OpenAI scaled back its original Instant Checkout in early 2026.
- Tracking: orders show up in your admin tagged with a ChatGPT referral. Find them under Analytics, then Reports, then Total sales by referrer.
Google AI Mode and Gemini
- Protocol: the Universal Commerce Protocol (UCP), co-developed by Shopify and Google and backed by Walmart, Target, Etsy, American Express, Mastercard, Stripe, and Visa.
- Where checkout happens: embedded right inside Google. The buyer never leaves AI Mode or the Gemini app.
- The trade-off: less friction, and likely the highest-conversion channel because the shopping intent is so strong, but you give up some checkout customization.
- Bonus: some merchants get access to Direct Offers, exclusive deals surfaced directly inside AI conversations.
Microsoft Copilot
- Where checkout happens: embedded Shopify-powered checkout inside Copilot, branded as Copilot Checkout.
- The audience: Copilot skews toward a productivity and enterprise crowd, a different buyer than ChatGPT’s general consumer base. Worth knowing when you think about which products to push.
- Fees: not formally confirmed, with no reports of fees beyond standard processing.
Perplexity (this one is different, read closely)
- How buying works: Perplexity’s in-chat purchase is called Buy with Pro, and it runs on PayPal, not on your Shopify checkout.
- How to get in: the free Perplexity Merchant Program connects your catalog to Perplexity’s shopping index. US-based Shopify stores can auto-share product data. A Perplexity Pro subscriber then sees a Buy with Pro button and checks out using PayPal details already saved to their Perplexity account.
- The catch for merchants: because checkout runs through PayPal Instant Buy, those orders are managed in PayPal, not in your Shopify admin. Plan your fulfillment and reporting around that.
- The upside: zero fees, zero commission, zero listing cost. That zero-fee model is Perplexity’s edge over ACP.
| Platform | Protocol | Checkout happens | In-chat? | Fees | Orders land in |
|---|---|---|---|---|---|
| ChatGPT | ACP (OpenAI + Stripe) | Your store | No | Standard Shopify rates | Shopify admin |
| Google AI Mode / Gemini | UCP (Shopify + Google) | Inside Google | Yes | No extra Google fee for now | Shopify admin |
| Microsoft Copilot | Copilot Checkout | Inside Copilot | Yes | Standard processing | Shopify admin |
| Perplexity | Buy with Pro (PayPal) | Inside Perplexity | Yes | Zero fees | PayPal |

Being in the Catalog vs Actually Getting Recommended
Here is the trap most stores are about to fall into. By default, your products are in the Shopify Catalog. That gets you listed. It does not get you recommended. Those are two completely different things, and the gap between them is the whole game.
An AI agent does not show every matching product. It reads the shopper’s intent and compares attributes like price, features, reviews, and availability to pick the best fit. The Shopify Catalog uses specialized AI models to categorize and standardize your data, which means being optimized is simply structuring your data so that the AI’s categorization and fact-extraction land on you instead of your competitor.
If that sounds familiar, it should. This is answer engine optimization pointed at the checkout. Same game as getting cited in AI search, just for products. Which brings us to the work.
How to Turn On Agentic Storefronts (Step by Step)
Good news first: if you are on Shopify, this is probably already on. But here is how to find it, check it, and control it.
Step 1. In your Shopify admin, go to Settings, then Sales channels.
Step 2. Find the Agentic Storefronts section. If it is still rolling out to your store, you can sign up at shopify.com/chatgpt.
Step 3. Review your ChatGPT settings. There is no opt-in toggle for ChatGPT discovery, it is automatic. This is the channel your products are already flowing into.
Step 4. For Google AI Mode, Gemini, and Copilot, set the “Allow customers to purchase directly in this sales channel” toggle on or off per channel. This decides whether buyers check out inside the AI app or get sent to your store.
Step 5. Use product visibility settings to hide any specific products you do not want surfaced to AI shoppers.
Step 6. Confirm attribution is flowing. Place a test order, or after your first AI order check Analytics, then Reports, then Total sales by referrer.
That is the entire setup. No app, no API, no developer. Which is exactly why the setup is not where you win. The next section is.
How to Optimize Your Products for AI Discovery
This is where the real work lives, and where most stores will lose. Being in the catalog is table stakes. Getting recommended is won field by field. Here is exactly what to fix, and why each one matters to the AI making the pick.
1. Write product titles an AI can actually parse
AI agents read titles semantically, not as keyword soup. A strong title answers five questions: what is it, what is it for, who is it for, what makes it different, and which variant is this.
Weak: The Alpine Pro
Strong: Alpine Pro Waterproof Hiking Boot, Men’s, Vibram Sole, Mid-Height
Use this formula:
[Brand] [Core Product] [Key Differentiator], [Target User], [Material or Spec], [Variant]
When a shopper asks ChatGPT for “the best waterproof hiking boot for wide feet,” the product whose title and data answer that question wins the recommendation before the shopper ever sees a store.
2. Write descriptions that answer the follow-up questions
AI agents pull facts from your descriptions to answer questions like “is this machine washable?” or “does this fit wide feet?” Pack in materials, dimensions, use cases, care instructions, compatibility, and what makes the product different. Specificity reduces ambiguity, and when the AI has to choose between three similar products, the one with explicit facts wins.
3. Use specific categories and tags
“Clothing” tells the AI nothing. “Women’s Activewear, Running Shorts, Lined” tells it use case, audience, and variant. Shopify’s standard categories run through specialized AI models, but accurate, specific input makes that categorization faster and more accurate.
4. Give it real images
AI platforms display your product images directly in the conversation. A single flat-lay loses to multiple angles plus lifestyle context with a model showing scale and fit. When a shopper wonders “will this actually work for me,” your imagery is part of the answer.
5. Keep pricing and inventory synced in real time
If a product is out of stock or the price is wrong, the AI either skips you or shows bad data. Both kill the sale and the trust. Your Shopify pricing and inventory have to be accurate continuously, not “mostly.”
6. Stack up real reviews
AI agents weigh review data when choosing what to recommend. Between two similar products, the one with 500 reviews at 4.8 stars beats the one with 20. High ratings and high volume are a trust signal that lowers the AI’s uncertainty about recommending you.
7. Set up the Knowledge Base app
This is the one most stores skip. Shopify’s Knowledge Base app lets you define your brand voice, FAQ answers, and return policy. When a shopper asks “what is your return policy?” inside ChatGPT, the AI pulls from your Knowledge Base instead of a generic default. It is how you control how your brand shows up in the conversation.
The AI Discovery Readiness Checklist
Copy this, paste it into your project tracker, and work down the list.
[ ] Product titles follow: [Brand] [Product] [Differentiator], [User], [Spec], [Variant]
[ ] Descriptions include materials, dimensions, use cases, care, compatibility
[ ] Categories use Shopify standard taxonomy, as specific as possible
[ ] Every product has multiple images plus lifestyle context
[ ] Pricing and inventory sync in real time
[ ] Review volume and rating are healthy on your hero products
[ ] Knowledge Base app set up with brand voice, FAQs, return policy
[ ] Per-channel direct-checkout toggles set the way you want them
[ ] Attribution confirmed in Analytics, Reports, Total sales by referrer
Why Your Storefront Matters More Now, Not Less
You might think AI doing the discovery makes your store less important. The opposite is true, and this is the part most people miss.
When someone buys through ChatGPT, they had zero prior relationship with you. They did not see your homepage, your story, or your Instagram. Their entire experience of your brand so far is a product card in a chat. And in a fast AI conversation, decisions happen faster, which means mistakes happen faster too. Someone can buy the wrong size in seconds.
That makes everything after the sale your whole relationship. The order confirmation email is often the first real touchpoint with your actual brand. A generic Shopify template wastes it. A confirmation with your voice, product care tips, and a clear way to fix or manage the order builds loyalty from second one. Because here is the hard truth: an AI-discovered customer has zero loyalty yet, and if something is wrong and self-service is not there, they will just request a refund. Your checkout and your post-purchase experience are no longer the back end. They are the brand.
Frequently Asked Questions
Does selling on ChatGPT cost extra?
No extra transaction fee in the live ChatGPT implementation. You pay standard Shopify processing rates. Perplexity’s merchant program is free with no commission. A few channels’ fee terms are still settling, so check your Shopify admin for the current state.
Do I have to do anything to turn it on?
If you are on an eligible Shopify plan, ChatGPT discovery is on by default. You manage the other channels’ direct-checkout toggles in Settings, then Sales channels.
Where do my orders show up?
ChatGPT, Google, and Copilot orders land in your Shopify admin with referral attribution. Perplexity Buy with Pro orders are handled through PayPal, so check there for those.
Can customers buy subscriptions or bundles this way?
Not directly yet. AI agents can recommend them, but the customer completes those purchases on your store.
Will my checkout customizations still work?
ChatGPT sends buyers to your real checkout, so yes. Some embedded channels do not support heavy checkout blocks, so test it if you run a customized Shopify Plus checkout.
Do discount codes work?
Only if you have configured a discount code field in the agentic checkout settings. Default discounts do not auto-apply.
Can I opt out of being shown?
You cannot fully opt out of ChatGPT discovery through the toggle, since it is a discovery channel. To hide specific products, use your product visibility settings.
What about returns?
Self-serve returns are not available through agentic storefronts yet. Returns run through your normal process, which is exactly why your post-purchase flow matters so much.
Do It Yourself, or Have It Done for You
Everything above is something you can do this week with no developer. Turn on the channels, fix your product data field by field, set up your Knowledge Base, and confirm your tracking. That alone will put you ahead of most stores, which are getting AI traffic right now and fumbling it.
If you would rather have it built and optimized for you, that is the kind of work I do. Reach out here and we can talk through your store. Either way, the window where this is still an edge is open right now. The stores that structure their data first are the ones the AI keeps recommending.